NOTES
5122
Even changing the nature of the ion produced no electrode action in liquid ammonia. In aqueous solutions at a high pH other ions, most frequently sodium ions, will take part in the transport of current, causing the well known "sodium error" of glass electrodes. The presence of a large excess of sodium ion in liquid ammonia solution did not produce an electrode response with any of the glasses tested. Ammonium ions in liquid ammonia are the counterpart of hydronium ions in water: they change colors of pH indicators, catalyze solvolysis reactions, are displaced by certain metals and set up potentials with a hydrogen electrode.g Glass electrode action might therefore also be expected in liquid ammonia solutions. Nevertheless, there are differences: the heat of solvation of the proton is 17 kcal./mole lower in liquid ammonia than in water, indiating a more stable system in ammonia. The low mobility of ammonium ions in ammonia as compared with hydronium ions shows that there is no "Grotthuss" conduction in liquid ammonia. Though complete exchange exists between NH4+ and NH3,12 the dissociation of NH4+ into H + and NHs is very slight. Even in water ammonium ions have not been found to cause glass electrode "errors'' such as sodium ions. In terms of Dole's statistical treatment of glass electrode action,la the ammonium ion does not appear to have sufficient energy to exchange with the proton on the glass surface. Therefore, it seems plausible that the ammonium ions in liquid ammonia may not be in equilibrium with ions on the surface of the electrode. Experimental
The Cell.-A cylindrical cell of 70 mm. diameter provided with a large ground glass joint and several smaller joints in the head for the lead-plated platinum reference electrode and glass electrode was used. The inside of the glass electrode was also filled with liquid ammbnia and was provided with another reference electrode. The cell was cooled by immersion in a Dewar flask in acetone and Dry Ice. Measuring Equipment.-A Beckman model G PH meter mas used for the purpose of detecting PH response. A Precision multirange test meter was used for all current measurements. With this meter 0.1 microampere can be detected. A 2400-volt d.c. power supply was placed in series with the cell and the ammeter. When PH measurements were attempted the power supply was removed and the pH meter inserted. p H Sensitive Glasses.-Initial studies were confined entirely t o Corning 015 type glass. In all cases the pH sensitivity of the membranes was first checked in water. The membranes were then subjected to evacuation followed by long periods of immersion in liquid ammonia. This procedure was adopted t o ensure complete replacement of the water in the membrane by liquid ammonia. Membranes composed of PH sensitive glasses of different compositions were similarly tested, L e . , Beckman General Purpose glass and Beckman type E. Measurements and Results.-A power supply of 300 and 2400 v. d.c., respectively, the cell and the microammeter were connected in series. The resistance of the glass membrane was calculated from Ohm's law. Measurements were made in aqueous solution between 6" and 62" for Corning 015 and Beckman general purpose glass and between 26" and 30' for the Beckman type E glass. The resistance for Corning 015 glass varied between 21 megohms at 62" t o 1300 i 200 megohms at 6". For Beckman General Purpose glass the resistance varied between 9 megohms at 50" and 600 megohms a t 7'. (12) C. J. Nyman, S. C. Fung and H. W. Dodgen, THIS JOURNAL, 72, 1033 (1950). (13) M. Dole, J . Ckem. P h y s . , 2, 862 (1934).
VOl. 75
Similar measurements were performed with aqueous alcohol solutions containing 80% of alcohol and containing a small amount of hydrochloric acid. Measurements could now be carried out t o -33", giving a resistance of 2 X IO3 megohms for Corning 016 glass and 1.2 X 10' megohms for Beckman General Purpose glass. In liquid ammonia solutions no current could be observed in the temperature range investigated, -35 to -lo. BOSTON UNIVERSITY DEPARTMENT OF CHEMISTRY BOSTON15, MASS.
Sulfur Analogs of Isoquinolines and p-Carbolines. ,111. The Pomeranz-kitsch Reaction BY WERNER HERZAND LIN TSAI RECEIVED JUNE 8, 1953
In a previous paper' it was demonstrated that isosteres of substituted isoquinolines may be prepared by applying the Bischler-Napieralski reaction to derivatives of thiophene. This work is being extended in our laboratory. It also seemed desirable to investigate the feasibility of synthesizing the parent substances, thieno [2,3-c]- (I) and thieno[3,2-c]pyridine (11) by meam of the PomeranzFritsch reaction2 inasmuch as 2- and 3-thiophenealdehyde are readily available. A comparison of the properties of these substances with those of isoquinoline would be of great interest.
r\s/1 icq
/I
I
---AN
1s' ' I1
Although the condensation of the aldehydes with aminoacetal proceeded with great ease, cyclization of the resulting Schiff bases under the usual conditions gave only small yields of the desired thienopyridines I and 11. This seems to be due largely to resinification under the influence of sulfuric acid at the temperatures necessary to effect cyclization, Slightly better yields were achieved by using a mixture of polyphosphoric acid and phosphorus oxychloride as cyclizing agent. To our knowledge this medium has not previously been used in such reactions. In order to supplement our earlier work on the preparation of sulfur analogs of 8-carb01ine.s~studies were also carried out which had as their aim the preparation of thianaphtheno [3,2-c]- (111) and thianaphtheno [2,3-c]pyridine (IV) from the aminoacetals derived from 2- and 3-thianaphthenealdehyde. III' and IV6 have since been prepared by different, somewhat less convenient, routes. The cyclization did not proceed in sulfuric acid solution, but using polyphosphoric acid-phosphorus oxychloride the
I11
IV
(1) W. Hers, THISJOURNAL. 78, 351 (1951). (2) W. J. Censlex. Organic Reactions, 6, 191 (1051). (3) w. HWZ, TEWJOURNAL, 79,4999 (1950). (4) H. Kotake and T. Sakan, J . Insl. Polykch. Osaka Cify Uniu., Scr. C., 8 , No. 1.25 (1951); C.A., 46, 6121 (1952). (5) D. B. Capps and C. S. Hamilton, TEIS JOURNAL, 76, 697 (1963).
Oct. 20, 1953
NOTES
5123
removal of ether was distilled or dissolved in a little ethanol and treated with 5 ml. of a saturated solution of picric acid in ethanol. The picrate was collected after chilling and recrystallized from ethanol. (B) Polyphosphoric Acid Method.-A solution of 1 g. of phosphorus oxychloride in 30 g. of polyphosphoric acid was heated to the desired temperature in a n oil-bath, the aminoacetal added with stirring, the mixture stirred for 5 to 30 minutes and then poured into ice-water. The aqueous solution was extracted with ether, made basic, saturated with ammonium sulfate and again extracted with ether. The ethereal solution was dried and the solvent removed under reduced pressure. The residue was sublimed in vacuo. Thieno(2,3-c)pyridine (I).-Procedure A yielded 102 mg. of a picrate which decomposed a t 207.5-208.5'. Anal. Calcd. for C18H8N4SO7:C, 42.86; H, 2.21; N, 15.38. Found: C,43.42; H, 2.26; N, 15.26. Procedure B, using 10 g. of Schiff base in 200 g. of polyphosphoric acid and 10 g. of phosphorus oxychloride a t 120' for 30 minutes, yielded 225 mg. of crude thieno(2,3-c)pyridine by fractional sublimation in vacuo of the basic residue. Two sublimations a t a bath temperature of 70' (0.5 mm.) furnished colorless crystals, m.p. 54-55', Anal. Calcd. for C,H6NS: C, 62.19; H, 3.73. Found: C, 62.39; H, 3.90. After collection of the first fraction, the bath temperature was raised t o 130-140" and 90 mg. of a higher-melting solid, n1.p. 142-144', was obtained. This material has not yet been identified. Thieno(3,Z-c)pyridine (II).-Two and three-tenths grams of the Schiff base in 15 E. of DolvDhosDhoric acid and 1 E. of phosphorus oxychlorideat 120-i30° for 10 minutes yidded 150 mg. of crude base. Resublimation a t room temperaExperimental' ture (0.25 mm.) furnished an analytical sample, m.p. 42Diethyl 2-Thenylideneaminoaceta1.-A mixture of 11.2 43O. g. of 2-thiophenealdehyde and 14 g. of aminoacetal was Anal. Calcd. for C~HSNS:C, 62.19; H, 3.73; h', 10.36. heated on the steam-bath. After two hours ether was added and the mixture was dried over sodium sulfate. Re- Found: C, 61.72; H, 3.69; N, 10.01. moval of ether followed by fractionation in ~ C U Oyielded The picrate melted a t 224.5'. 19.6 g. (86%) of condensation product, b.p. 125-128' (2.5 Anal. Calcd. for CIZH.S?~~OB: C, 42.56; H, 2.21; N, mm.). The analytical sample boiled a t 115-118' (1.9 mm.), 15.38. Found: C, 42.88; H, 2.27; N, 14.9. nlaD 1.5231. Procedure A yielded 201 mg. of picrate. Anal. Calcd. for CIIHIINSOI: C, 58.12; H, 7.54. Thianaphtheno(2,3-~)pyridine (III).-From 1.3 g. of the Found: C, 58.16; H, 7.55. Schiff base in 31 g. of polyphosphoric acid and 1 g. of phosDiethyl 3-Thenylideneaminoacetal.-Condensation of phorus oxychloride a t 150-160' for 20 minutes there was similar quantities of 3-thi~phenealdehyde~ and amino acetal obtained 240 mg. of the base. This was purified first by yielded 20.5 g. of product, b.p. 120-121' (1.8 mm.). The recrystallization from ether-petroleum ether and finally by sublimation a t 0.2 mm., m.p. 96-98' (lita498-99'), The analytical sample boiled at 128' (3 mm.), n% 1.5179. Anal. Calcd. for CI1&h'S02: C, 58.12; H, 7.54. picrate decomposed a t 256' (lit.4 258-260'). Xo cyclization was effected by means of 96% sulfuric Found: C, 58.21; H, 7.39. a t 5' for 10 hours as well as a t 100' for two hours with Diethyl 2-ThianaphthenylideneaminoacetaL-The yield of aacid mixture of phosphorus oxychloride and sulfuric acid. condensation product from 14.4 g. of 2-thianaphthenealdeThianaphtheno(3,O-c)pyridine (IV).-From 1.6 g. of the hydeg and 12 g. of aminoacetal was 19 g. (78%), b.p. 155- Schiff in 30 g. of polyphosphoric acid and 1 g. of phos161' (0.2 mm.). The liquid solidified upon standing and phorusbase oxychloride a t 90-100° for 20 minutes there was obwas sublimed for analysis at a bath temperature of 110- tained 0.2 g. of the base. Recrystallization from petroleum 125' (0.5 mm.), m.p. 45-46'. $her furnished colorless needles, m.p. 71-71.5' (1it.I 69Anal. Calcd. for CIIHIDO~NS:C, 64.95; H, 6.90; N, (0.5'). The picrate melted a t 248". 5.05. Found: C,65.05; H , 7.17; N, 5.15. Anal. Calcd. for CI~HIOO,N,S:N, 13.39. Found: N, Diethyl 3-Thianaphthenylideneaminoacetal.-The yield 13.6. of product, b.p. 155-165' (1.5 mm.), was 20.4 g. (83%). OF CHEMISTRY The analytical sample boiled a t 162-165' (1.5 mm.), nZ31) DEPARTMENT THEFLORIDA STATE UNIVERSITY 1.5711. FLORIDA Anal. Calcd. for CIJ-LONSOI: C, 64.95; H, 6.90. TALLAHASSEE, Found: C, 64.93; H, 6.82. Cyclizations. (A) Sulfuric Acid Method.-To 6 g. of ice-cold sulfuric acid was gradually added 3 g. of the amino- Preparation of Normal and Secondary Butyl Hydroperoxides acetal, then another 9 g. of sulfuric acid and 2 g. of phosphorus oxychloride. The mixture was kept at 160' for BY E. G. LINDSTROM 1.5 hours, in the course of which it resinified, cooled and diluted with water. The filtered solution was extracted with RECEIVED JULY 6, 1953 ether (this fraction usually yielded aldehyde formed by hydrolytic cleavage of the acetal), made basic and again exThe classical method of preparing methyl and tracted with ether. The basic fraction after drying and ethyl hydroperoxides is the reaction of the appropri-
yields of thianaphthenopyridines were 18 and 12%. This suggests that polyphosphoric acid may be a useful cyclizing agent for other syntheses using the Pomeranz-Fritsch method. The ultraviolet spectrum of I in isooctane solution (maxima a t 223, 227, shoulder a t 242, maxima a t 293, 299 and 304 mp; corresponding log E'S 4.24, 4.23, 3.68, 3.88, 3.76) exhibits marked resemblance to that of isoquinoline,6 but that of I1 (maxima a t 222, 238, 236, shoulder a t 274, maxima a t 281, 291 and 337 p, log e's4.55, 3.58, 3.63, 3.45, 3.43, 3.30, 2.35) does not possess this similarity in the longer wave length region. It may be that in the spectrum of I1 the 290-310 mp band of I is suffciently reduced in intensity and displaced toward shorter wave lengths to give rise to the less well pronounced maxima in the region 270-295 mp. A somewhat similar situation is found in the spectra of the two isomeric thianaphthenopyridines I11 (maxima a t 232, 272, 280, 328 and 338 mp, log E'S 4.61, 3.88, 4.03, 3.53 and 3.60) and IV (maxima a t 230, 252, 305 and 316 mp, log 6's 4.70, 4.33, 3.15 and 3.21). Acknowledgment.-This work was supported in part by a grant from the U. S. Public Health Service, Department of Health, Education and Welfare.
(6) R . A. Friedel and M. Orchin, "Ultraviolet Spectra of Aromatic Compounds," John Wiley and Sons, Inc., New York, N. Y., 1951. (7) Melting points and boiling points are uncorrected. Analyses by Clark Microanalytical Laboratory, Urbana, Illinois, and Drs. Weiler and Straws, Oxford. (8) E. Campaigne and W. LeSuer, THISJOURNAL, 70, 1555 (1948). (9) D. A. Shirley and M. J. Danzig, i b i d . , 74, 2035 ( l @ J 2 ) .
ate alkyl sulfate with hydrogen peroxide in concentrated aqueous potassium hydroxide solution.' It appears, however, that the method is not applicable to higher members of the series. Medvedev2 was (1) A. Baeyer and V. Villiger, Bcr., 34, 738 (1901). (2) S. S. hledvedev and E. N. Alekseeva, ibad., 66B, 133 (3932).
